Prognostic Value of Different Versions of the Model for End-Stage Liver Disease Score in Patients Undergoing Percutaneous Coronary Intervention

Abstract

The model for end-stage liver disease (MELD) score, which can reflect liver and renal function, is associated with poor prognosis. However, the prognostic performance of the modified MELD score in patients undergoing elective percutaneous coronary intervention (PCI) has not been fully evaluated and compared. This study retrospectively enrolled 5324 patients. During a median follow-up of 2.85 years, 412 patients died. Time-dependent receiver operating characteristic curves at 3 years indicated that the MELD including albumin (MELD-Albumin) score had the highest prognostic performance (AUC = .721) than the MELD score (AUC = .630), the MELD excluding the international normalized ratio (MELD-XI) score (AUC = .606), and the MELD including sodium (MELD-Na) score (AUC = .656) (all P < .001). The MELD-Albumin score, the MELD score, and the MELD-Na score were independent predictors of long-term mortality; however, the MELD-XI score was not when treated as a categorical variable ( P = .254). Adding the MELD-Albumin score to the model of clinical risk factors could improve the prognostic performance. For the subgroup analysis, the association between the MELD-Albumin score and long-term mortality was more pronounced in patients ≤75 years (interaction P value = .005). The MELD-Albumin score showed the strongest prognostic performance than the other versions of the MELD score in patients undergoing elective PCI.